Bank Check · Day 4 of 6. An investor's guide to the cost-to-income ratio: what it measures, why different kinds of banks sit at different levels, and how to tell real efficiency from one offs.
In the quarter to 30 September 2026, one bank in our comparison spent about ₹38 for every ₹100 of operating income. Another spent ₹62. The second is not automatically the worse-run bank. Here's how to read the number, and what it can and can't tell you.
What cost-to-income measures
A bank earns net interest income and fees. To earn them, it pays for staff, branches, technology and administration. The cost-to-income ratio shows how much of every rupee earned goes on running the bank, before bad loans and tax. A 45% ratio means roughly ₹45 spent for every ₹100 of operating income.
Read four numbers together:
- Cost-to-income = Operating expenses ÷ Operating income, where operating income is net interest income plus other income. How lean the bank is.
- Core cost-to-income = the same, with treasury gains and losses (on the bank's bond holdings) taken out. Whether the efficiency is real.
- The "jaws" = Income growth − Cost growth. Positive jaws mean income is growing faster than costs, so the ratio should fall.
- Cost-to-assets = Operating expenses ÷ Average assets. A fairer comparison when two banks' margins differ a lot.
A lower ratio leaves more of each rupee to absorb bad loans and turn into profit. But the ratio is a fraction, so it can move because the top changed (costs) or because the bottom changed (income).
Cost-to-income across bank categories, Q2 FY27
Across the 10 banks in our series, the ratio ran from about 38% to 62%, and it clustered by category.

Cost-to-income as reported for Q2 FY27, 10 banks · sources at the end
Large private banks benefit from scale and fee income. Public sector banks carry wage revisions, pension liabilities and a large branch network. Small finance banks lend in small tickets, which costs more to service. Five of the six banks with a year-ago figure reported a lower ratio than a year earlier; Canara Bank's edged up.
The securities quoted are for illustration only and are not recommendatory
What actually moved the number
A change in the ratio has one of three causes: income moved, a one-time cost landed, or the bank genuinely got leaner. One bank per category, Q2 FY27:
- Public sector : income moved. Treasury income fell to ₹753 crore from ₹1,769 crore a year earlier, which pushes the ratio up. Its H1 ratio still improved, to 50.38% from 53.3%, as core operating profit grew about 29%. The bank is targeting below 50% this year and about 45% by FY29.
- Large private : a one-time cost. The ratio rose to 49.4% from 45.5% the quarter before. Opex grew 9% from a year ago and included a one-time ₹250 crore cost tied to the new labor codes, plus the cost of raising FCNR(B) deposits.
- Mid-size private : real efficiency. Operating expenses fell about 2% while total income rose 12%. Those positive jaws took the ratio to 57.8% from 66.2% a year earlier.
- Small finance : the business model. A 62% ratio reflects small-ticket lending that needs more branches and staff per rupee lent. A NIM of 8.4% is what pays for it.
Before reading a jump or a fall, ask which side of the fraction moved, and whether it will repeat.
The securities quoted are for illustration only and are not recommendatory.
Read it with NIM and credit cost
The three numbers from this series fit into one chain: NIM → minus bad loans → minus running costs → profit.
Take ₹100 of operating income. At a 40% cost-to-income ratio, ₹60 is left before provisions. At 60%, ₹40 is left. Whether ₹40 is enough depends on how much income the bank makes on its assets in the first place, and how much it then loses to bad loans (Day 3).
That is why a small finance bank's ratio should be read against its margin, not against a large private bank's ratio. AU Small Finance Bank, for example, reported a NIM of 5.8% and running costs of 3.9% of assets (excluding guarantee premium) in Q2 FY27. Its costs are high, but its margin is higher still. If the margin narrows while costs stay put, that cushion shrinks fast.
Six tests before you read a bank's cost-to-income ratio
- Core, not headline. Red flag: a ratio that improves only because of a one-off bond gain.
- The jaws. Red flag: costs outgrowing income for several quarters in a row.
- One-time costs. Red flag: "one-time" costs that show up again the next year.
- Like with like. Red flag: calling a small finance bank inefficient next to a large private bank.
- Against the margin. Red flag: a high ratio while NIM is shrinking.
- Investment or bloat. Red flag: costs rising while deposit and loan growth slow
(A learning aid, not a recommendation.)
The takeaway: efficiency is what each rupee of spending brings back
A lower ratio is not automatically a better bank. In FY26, public sector banks together posted record profits with a combined cost-to-income ratio of 49.67%, and the Finance Ministry has since pressed them for tighter cost control. Profit and cost discipline are related, but they are not the same thing.
Next in Bank Check: Day 5 covers ROA and ROE, which show whether all of this turns into returns on the bank's assets and on shareholders' money.
Frequently asked questions
What is a good cost-to-income ratio for an Indian bank?
There is no single good number. It depends on the bank's category and business model. In Q2 FY27, large private banks in our comparison ranged from about 38% to 49%, and small finance banks from about 58% to 62%. Compare a bank with its own peers and its own history.
Why do public sector banks have higher cost-to-income ratios?
They carry wage revisions, pension liabilities and a large branch network, and they serve financial-inclusion goals. A CareEdge report put the public sector banks' ratio at 53.9% in Q3 FY24, against 47.2% for private banks.
Can a bank's cost-to-income ratio rise without its costs rising?
Yes. The ratio also rises when income falls, for example when treasury income drops after bond losses. That is why it helps to check the core ratio, with treasury income taken out.




